DESCRIPTION: Genetic differences between individuals can greatly influence their susceptibility to disease. The information originating from the Human Genome Project (HGP), including the genome sequence and its annotation, together with projects such as the HapMap and the Human Cancer Genome Project (HCGP) have greatly accelerated our ability to find genetic variants and associate genes with a wide range of human diseases. Despite these advances, linking individual genes and their variations to disease remains a daunting challenge. Even where a causal variant has been identified, the biological insight that must precede a strategy for therapeutic intervention has generally been slow in coming. The primary reason for this is that the phenotypic effects of functional sequence variants are mediated by a dynamic network of gene products and metabolites, which exhibit emergent properties that cannot be understood one gene at a time. Our central hypothesis is that both human genetic variations and pathogens such as viruses influence local and global properties of networks to induce "disease states." Therefore, we propose a general approach to understanding cellular networks based on environmental and genetic perturbations of network structure and readout of the effects using interactome mapping, proteomic analysis, and transcriptional profiling. We have chosen a defined model system with a variety of disease outcomes: viral infection. We will explore the concept that one must understand changes in complex cellular networks to fully understand the link between genotype, environment, and phenotype. We will integrate observations from network-level perturbations caused by particular viruses together with genome-wide human variation datasets for related human diseases with the goal of developing general principles for data integration and network prediction, instantiation of these in open-source software tools, and development of testable hypotheses that can be used to assess the value of our methods. Our plans to achieve these goals are summarized in the following specific aims: 1. Profile all viral-host protein-protein interactions for a group of viruses with related biological properties. 2. Profile the perturbations that viral proteins induce on the transcriptome of their host cells. 3. Combine the resulting interaction and perturbation data to derive cellular network-based models. 4. Use the developed models to interpret genome-wide genetic variations observed in human disease, 5. Integrate the bioinformatics resources developed by the various CCSG members within a Bioinformatics Core for data management and dissemination. 6. Building on existing education and outreach programs, we plan to develop a genomic and network centered educational program, with particular emphasis on providing access for underrepresented minorities to internships, workshop, and scientific meetings.